Thanks Martin, for the info … now I figured out an additional method that works for manual modes.
A third source into the mixers:
S: 3 6 9000 9000 0 -10000 10000
S: 1 4 0 -20000 10000 -10000 10000
S: 0 2 8000 8000 0 -10000 10000
Then RcAux can be used as a kind of tilt trim. And with a mixer in the tx it can be assigned to the pitch stick that brings the desired mode back. And to control the amount of hanging tail in assisted modes as well.
But how can I get rid of the 2150us limit?
“pwm max -c 56 -p 2450” and “pwm min -c 56 -p 650” result in the desired min pwm but max pwm is limited at about 2153us
2450us would be the perfect value to follow the full path of the Convergence´s mechanical servo system (and a way to use PX4 for my Taileron Vtol project)
I tried the tuning, its better but still unstable after transision back to tricopter mode. Now, it destroyed after the 14th flight… im building the new one now…
I think it´s absolutely essential for full control and fun with the Convergence to have pwm signals that are only a small bit towards the original E-flite controllers values.
The original E-flite fc covers a (meassured) pwm range of 483 to 2560 us for the tilt rotor servos (ailerons 722 to 2322 us) thus I think a full safe (incl. symmetrical yaw) control of the Convergence with the PX4 default of 1000 to 2000 us is not really possible !!
I know with pwm rates of 400Hz for the esc at 2350 us there are only 150us pause time left. Escs mostly don´t need more than 1950 us.
But the servo outputs work at 50 hz where only a range of 500 … 2500us would fit for most vtol purposes. The low limits can be successfully set with PX4 but currently there is a upper limit at 2150us.
@taileron if you want to create an issue we can discuss and experiment with increasing the limit (or ignoring it). At the moment I don’t know if anything else in PX4 will prevent us from actually using that range.
It does, I’ve been flying the stock Convergence and its great fun.
Stock if you pitch down it mixes 30 to 40 degrees of forward pitch on the two motors. This is similar to the first stage of transition. With the tilt forward flight is quite impressive and wind penetration is very good. Without this mix the tail rotor isn’t powerful enough to give good forward flight. In wind it simply doesn’t work.
Here’s a video I made of the forward flight in hover mode.
Now it flies absolutely with great fun with a Pixracer and tiltforward like the stock controller I used the mixer below - it needs two additional tx mixers to redirect the pitch stick input to an aux channel that controls the amount of tilt in stability mode for position mode it needs both pitch and aux:
Only the tilt backward amount is currently much smaller than with the stock controller cause a pwm limit at 2150uS of the PX4 firmware limits the right tilt servo before the desired value
@AlexandreBorowczyk it´s not necessary especially ch6 - PX4 control group 3 is manual passthrough and also can be used as mixersource that afterwards can be assigned to any free ch of the tx. (PX4 parameter: RC_MAP_AUX3 = desired tx ch)
Finally there are to do mixers on the tx that temporarly send this desired ch instead of the pitch channel by moving the pitch stick.This mixers must be enabled by a tx switch to select between tilt and standard mode.
A mix pitch to pitch with -100% disables pitch output at all then pitch can be mixed to the desired tx ch
Hello Convergence flyers.
I just outfitted a Convergence with a Pixhawk and flew on Saturday. I was quite fun and worked with no problems. I’m using it as a testbed for tiltrotor vtol ideas I have in my head.
Some things I noticed.
Pixhawk works fine. Just remove it from the case and be creative on fitting everything in there.
I see someone else found this issue above. The wing ESC wires had black/white wires reversed. I had to open the wing pods and re-pin the connectors. Now motors work fine. (E-flite quality issue that I reported to them).
Help When I enabled position mode, the aircraft wandered around, barely controllable. When I enable HOLD mode the aircraft stuck, rock solid. I’ll pull my last dataflash log and upload it. But does anyone know why Position hold doesn’t hold position?
From my experience with the Convergence, pitching the aircraft doesn’t give enough velocity authority (due to the air frame aerodynamics) for the position controller to do the job. You need to add some rotor tilting action in the mixer and then adjuster some of the controller gains.
Hi, I’m also studying VTOL. I use Fun Cub Quad VTOL.I met a problem, but in the process of flying four rotor to fixed wing, switch mode to the fixed wing, four rotor however, did not stop, and fixed wing throttle is not controlled by me, their growth up.I hope you can help me.
Hi Dagar,
My sticks were centered at the beginning but I quickly used them to adjust the position drift that I observed. I notice in the plots, the setpoint changes once engaged (and then more when I correct with sticks).
This is the log with POS control and drift. https://logs.px4.io/plot_app?log=69412547-5d33-4e04-8d0e-6238dfcb2366
